Tag: multi-scale feature extraction

A compact residual network called MSERNet removes JPEG compression artifacts from ovarian and carotid ultrasound images while preserving tissue textures ...

A new 3D Mamba-based model segments brain tumors with consistent accuracy gains and a 15x efficiency improvement over existing state-of-the-art ...

A Chinese research team reports a new asymmetric U-Net with hybrid attention and residual dilated convolution that achieved Dice scores ...

A new YOLOv8-based algorithm called DARF improves visible-infrared object detection by enhancing shallow features and adaptively routing cross-modal information, achieving ...

A sine-guided Transformer–CNN network significantly outperformed 21 comparison methods on three benchmark datasets for salient-object detection in optical remote-sensing imagery.

An enhanced YOLOv11 model developed for printed circuit board inspection achieved mean average precision of 96.1% at an intersection-over-union threshold ...

Researchers at Hohai University have developed ShuffleNet-MSAA, a lightweight AI system that uses adaptive attention and sonar imaging to detect ...